Physiology and Pathology of Calcium Signaling in the Brain

Abstract

Calcium (Ca²⁺) plays fundamental and diversified roles in neuronal plasticity. As second messenger of many signaling pathways, Ca²⁺ as been shown to regulate neuronal gene expression, energy production, membrane excitability, synaptogenesis, synaptic transmission, and other processes underlying learning and memory and cell survival. The flexibility of Ca²⁺ signaling is achieved by modifying cytosolic Ca²⁺ concentrations via regulated opening of plasma membrane and subcellular Ca²⁺ sensitive channels. The spatiotemporal patterns of intracellular Ca²⁺ signals, and the ultimate cellular biological outcome, are also dependent upon termination mechanism, such as Ca²⁺ buffering, extracellular extrusion, and intra-organelle sequestration. Because of the central role played by Ca²⁺ in neuronal physiology, it is not surprising that even modest impairments of Ca²⁺ homeostasis result in profound functional alterations. Despite their heterogeneous etiology neurodegenerative disorders, as well as the healthy aging process, are all characterized by disruption of Ca²⁺ homeostasis and signaling. In this review we provide an overview of the main types of neuronal Ca²⁺ channels and their role in neuronal plasticity. We will also discuss the participation of Ca²⁺ signaling in neuronal aging and degeneration.

Introduction

Calcium (Ca²⁺) plays fundamental and diversified roles in neuronal physiology. For example, by regulating the release of neurotransmitters from the presynaptic terminals it influences both long-term potentiation (LTP; Grover and Teyler, 1990; Impey et al., 1996) and long-term depression (LTD; Bolshakov and Siegelbaum, 1994; Christie et al., 1996) forms of synaptic plasticity. As ubiquitous second messenger, Ca²⁺ has been shown to regulate gene expression (Berridge, 1998), membrane excitability (Sudhof, 2004), dendrite development (Lohmann and Wong, 2005; Redmond and Ghosh, 2005), synaptogenesis (Michaelsen and Lohmann, 2010), and many other processes contributing to the neuronal primary functions of information processing and memory storage (Berridge, 1998; Tanaka et al., 2008). The specificity of various biological outcomes is rendered possible by a complex protein network tightly regulating the amplitude, spatial, and temporal patterns of calcium movements through the neuronal cellular compartments. Under resting conditions neurons actively maintain a steep gradient between low intracellular free Ca²⁺ concentration [Ca²⁺]_i (0.1–0.5 μM) and high extracellular Ca²⁺ levels (∼1 mM). [Ca²⁺]_i changes result from influx into the cell regulated through the opening of voltage-dependent Ca²⁺ channels (VDCCs), N-Methyl-d-Aspartate (NMDA) receptors, or transient receptor potential (TRP) channels located in the plasma membrane. Additionally [Ca²⁺]_i levels can be increased via ryanodine receptors (RyRs) and inositol-1,4,5-triphosphate receptors (IP₃Rs) mediated release from endoplasmic reticulum intracellular Ca²⁺ stores, or by sodium-dependent Ca²⁺ (Na⁺/Ca²⁺ exchanger) efflux from mitochondria. The return to basal [Ca²⁺]_i is achieved by Ca²⁺ extrusion (i.e., plasma membrane Ca²⁺-ATPase, sodium calcium exchanger), binding to Ca²⁺ buffering proteins (e.g., calmodulin, calcineurin, calbindin, calretinin), or organelle uptake (i.e., sarcoplasmic–endoplasmic reticulum Ca²⁺–ATPase, mitochondrial uniporter; Carafoli, 1987; Miller, 1991; Zaidi et al., 2003; Pivovarova and Andrews, 2010).

The purpose of this review is to provide a brief overview of the main types of neuronal Ca²⁺ channels and their role in neuronal plasticity. We will also discuss the participation of altered Ca²⁺ homeostasis to the cognitive decline observed during aging and the neurodegenerative process.

Plasma Membrane Ca²⁺ Channels

Neuronal Ca²⁺ influx is primarily regulated through two types of plasma membrane Ca²⁺ channels: VDCCs and receptor-operated (ligand-gated) channels (ROCs).

Voltage-dependent Ca²⁺ channels

Voltage-dependent Ca²⁺ channelss transduce electrical signals into local intracellular Ca²⁺ transients regulating intracellular processes such as neurotransmission, enzyme activation, gene expression, and neurite outgrowth or retraction (Tsien et al., 1988; Catterall, 2011). They are composed by an α1_1–10 subunit forming the Ca²⁺ selective channel, and several accessory subunits, α2δ, β_1–4, and γ with anchorage, and regulatory functions. Based on their unique electrophysiological and pharmacological properties, and the type of α1 subunit VDCCs are divided into five classes: Ca_v1.1–Ca_v1.4 (L-type Ca²⁺ current), Ca_v2.1 (P/Q-type Ca²⁺ current), Ca_v2.2 (N-type Ca²⁺ current), Ca_v2.3 (R-type Ca²⁺ current), and Ca_v3.1–3.3 (T-type Ca²⁺ currents; Tsien et al., 1988; Catterall et al., 2005; Catterall, 2011).

In the mammalian brain Ca_v1.2 and Ca_v1.3 are the predominant forms of L-type Ca²⁺ channels. They are localized in the cell bodies and proximal dendrites, both at presynaptic as well as postsynaptic locations, forming small clusters on dendritic shafts and spines (Hell et al., 1993; Obermair et al., 2004) associated with regulatory proteins such as, protein kinase-A, A kinase anchoring proteins (AKAP15), calmodulin, and calcineurin (Davare et al., 2001). L-type channels have a high-voltage threshold for activation and are important for integrating synaptic inputs with the initiation of neurotransmitter release. Inward L-type Ca²⁺ currents can directly depolarize the membrane potential of neurons (Moyer and Disterhoft, 1994; Chan et al., 2007), or alternatively depress membrane excitability through coupling to Ca²⁺-activated K⁺ channels (Wisgirda and Dryer, 1994; Marrion and Tavalin, 1998). Because of their preferential cellular localization and interaction with Ca²⁺-binding regulatory proteins (they have been shown to regulate nuclear gene transcription (Dolmetsch, 2003; Oliveria et al., 2007). In addition, L-type channels are required for NMDA receptor-independent LTP at synapses between CA3 pyramidal neurons and mossy fibers, and between dentate gyrus and the basolateral nucleus of the amygdala (Grover and Teyler, 1990; Cavus and Teyler, 1996; Raymond and Redman, 2002, 2006; Niikura et al., 2004), spatial memory (Moosmang et al., 2005) and heterosynaptic plasticity (Lee et al., 2009; Rose et al., 2009). L-type Ca²⁺ currents leading to global dysregulated calcium signals have been implicated in aging and neurodegenerative disease (Chang et al., 2009). Enhanced activity of L-type Ca²⁺ channels is observed with neuronal aging (Thibault and Landfield, 1996) and their expression is increased in hippocampi of patients with Alzheimer’s disease (AD) compared to healthy subjects (Coon et al., 1999). Blockage of L-type currents has been shown to improve learning and memory in aged mice (Disterhoft and Oh, 2006), and in patients with dementia (Forette et al., 2002). In addition, inhibition of L-type channels reduces cell death in stroke (Korenkov et al., 2000) and Parkinson’s disease (Chan et al., 2007). In young neurons, sustained depolarization or activation of NMDA receptors reduces L-type Ca²⁺ currents by internalization of Ca_V1.2 channels trough a dynamin-dependent endocytosis mechanism, protecting the neurons from excitotoxic cell death (Green et al., 2007). It is thus possible that some of the impairments observed in neurons in aging and degenerative disorders are due to an altered regulation of Ca_V1.2 and Ca_v1.3 levels at the membrane.

Ca_v2.1 (P/Q-type Ca²⁺ current) and Ca_v2.2 (N-type Ca²⁺ current) channels are the predominant Ca²⁺-dependent pathway triggering fast release of neurotransmitters like glutamate, γ-aminobutyric acid (GABA), and acetylcholine (Olivera et al., 1994; Dunlap et al., 1995). Ca_v2 are high-voltage threshold activated channels localized at presynaptic terminals, dendrites and cell bodies (Westenbroek et al., 1992). Ca_v2 channels directly interact with the SNARE complex, formed by the vesicle-associated-v-SNARE protein synaptobrevin (VAMP/synaptobrevin) and two plasma-membrane-associated t-SNARE proteins, SNAP-25 and syntaxin-1 (Bajjalieh and Scheller, 1995; Sudhof, 2004). The interaction occurs at a specific site, synprint, in the large intracellular loop connecting domains II and III of the α1 subunit of Ca_v2 channels (Sheng et al., 1994; Retting et al., 1996). This association is Ca²⁺ dependent and regulated by protein phosphorylation of the synprint domain by protein kinase C (PKC) and Ca²⁺/calmodulin-dependent protein kinase II (CaMKII; Sheng et al., 1996; Yokoyama et al., 2005). The same site is also responsible for the binding to synaptic vesicle Ca²⁺-binding protein synaptotagmin (Charvin et al., 1997; Wiser et al., 1997). During depolarization the presynaptic Ca_v2.1 and Ca_v2.2 channels open allowing the formation of high Ca²⁺ concentration microdomains in proximity of the pore (Stanley, 1997). Ca²⁺ then binds to synaptotagmin and the SNARE complex, resulting in the fusion of the vesicular membrane with the plasma membrane, and release of the neurotransmitter (Catterall and Few, 2008). Inhibition of presynaptic P/Q-types and N-type currents with reduction of neurotransmitter release is typically produced by a positive shift in the voltage dependence and a slowing of channel activation (Bean, 1989). The Gβγ subunits released from receptor-coupled heteromeric G-proteins of the Gi/Go class are usually responsible for this inhibition by binding to the loop between domains I and II and the amino- and carboxy-terminal domains of Ca²⁺ channels (Herlitze et al., 1996; De Waard et al., 1997; Zamponi et al., 1997; Page et al., 1998; Li et al., 2004). Contrary to Ca_v2.1 and Ca_v2.2, Ca_v2.3 (R-type Ca²⁺ current) channels control transmitter release with a lower efficacy (Wu et al., 1998). This is probably due to the fact the although they are localized at the cell body, dendrites and presynaptic terminal, their position is further away from the release sites (Yokoyama et al., 1995; Day et al., 1996; Wu et al., 1999). Nevertheless, Ca_V2.3 still carries one-third of the total calcium current at presynaptic terminals during a presynaptic action potential (Wu et al., 1998). Dietrich et al. (2003) showed that Ca²⁺ influx through presynaptic Ca_v2.3 contributes to LTP without playing a role in the fast synaptic transmission or facilitation at specific hippocampal synapse terminals. It has been suggested that Ca_V2.3 currents may be involved in control of gene expression or dendritic excitability (Delmas et al., 2000). Pathophysiological changes of Ca_v2 channels have been associated with chronic disorders. Mutations in the CACNA₁A gene encoding the Ca_v2.1 channels have been identified in patients affected by epileptic seizures, episodic ataxia type-2, spinocerebellar ataxia type-6, and familial hemiplegic migraine type 1 (Catterall et al., 2005; Cain and Snutch, 2011).

Ca_v3.1–3.3 channels (T-type Ca²⁺ currents) are mainly distributed at cell bodies and dendrites of neurons in the olfactory bulb, amygdala, cerebellar cortex, hippocampus, thalamus, hypothalamus, and striatum (Talley et al., 1999). These channels play a critical role toward neuronal firing both in conducting Ca²⁺ during action potentials and in switching neurons between distinct rhythmic firing modes. T-types Ca²⁺ currents are activated at rather negative near resting membrane potentials (Huguenard, 1996). Specifically, they are activated during the initial depolarization phase although the highest conductance occur during the repolarization phase and return to resting membrane potential (McCobb and Beam, 1991). The Ca²⁺ entry through the Ca_v3.1–3.3 channels leads to depolarization of the membrane allowing the generation of low threshold spikes that trigger bursts of Na-dependent action potentials (Llinas, 1988). Depending on the specific channel subtype the time course of activation, inactivation, deactivation, and recovery from inactivation varies, resulting in unique biophysical properties and specific responses to action potential which can be extended during burst firing (Huguenard, 1996). Moreover the expression of individual or multiple Ca_v3 subtypes and splice variants results in a variety of different burst patterns (Cain and Snutch, 2010). Bursts discharges controlled by T-type Ca²⁺ currents occur during physiological and pathological forms of neuronal rhythmicity (Huguenard, 1996) such as slow sleep oscillations (<1 Hz; Crunelli et al., 2006), learning (Scotty et al., 2003) and hyper-synchronous oscillations during epilepsy (Zamponi et al., 2010). Mutations in the CACNA₁H gene encoding the Cav3.2 channels have been linked to epilepsy and autism spectrum disorders (Heron et al., 2004; Splawski et al., 2006).

Receptor-operated (ligand-gated) channels

Receptor-operated (ligand-gated) channels open in response to the binding of specific ligands, such as neurotransmitters to the extracellular domain of the receptor. The interaction causes a conformational change in the structure of the protein that leads to the opening of the channel pore and subsequent ion flux across the plasma membrane. Most ROCs are permeable to Ca²⁺ and represent an important mechanism for the generation of second messengers. Examples of ROCs include the glutamate receptors NMDA, α-amino-3-hydroxy-5-methylisoxazole-4-propionate acid (AMPA; Isaac et al., 2007; Paoletti, 2011), and kainate (KARs; Chittajallu et al., 1999), nicotinic acetylcholine receptors (nACh; Shen and Yakel, 2009), serotonin (5-HT₃) receptors (Yakel, 2000), and adenosine 5′-triphosphate (ATP) P2X receptors (Pankratov et al., 2008).

Glutamate receptors

Glutamate is the most abundant neurotransmitter in the central nervous system (CNS) playing an important role in neuronal physiology and pathology. At excitatory synapses glutamate release will primary activate the ionotropic receptors, NMDA, AMPA, and KARs allowing rapid influx of ions into the postsynaptic terminal.

NMDA receptors are permeable to Na⁺, which contributes to postsynaptic depolarization, and Ca²⁺, which generate Ca²⁺ transients and the ultimate intracellular physiological response. A unique feature of these receptors is that at resting potential the cation pore is blocked by extracellular Mg²⁺. Their activation thus relies upon glutamate binding to postsynaptic AMPA receptors and Na⁺ and Ca²⁺ influx to cause a partial membrane depolarization sufficient to lift the Mg²⁺ blockage. Molecular cloning studies have identified several variants of the NMDA receptor subunits, NR1_1–4a/b and NR2_A–D (Hollmann and Heinemann, 1994). Each channel consists of two NR1 subunits, which are essential for the assembly of functional NMDA receptors, and two NR2 subunits that confer a unique set of characteristic upon the resultant NMDA receptor (Kutsuwada et al., 1992; Monyer et al., 1992). The cytoplasmic C-terminals of the NR1 and NR2 subunits link the receptor to a large multi-protein complex. This interaction facilitates the localization of the receptor in specific areas, such as the postsynaptic density (PSD), and the connection to a variety of downstream signaling molecules (Traynelis et al., 2010). In particular, NMDA-dependent Ca²⁺ influxes are known to regulate CREB-dependent gene transcription (Hardingham et al., 1999, 2001a,b; Impey and Goodman, 2001; Wu et al., 2001) which has considerable physiological relevance for the establishment of long-term synaptic plasticity and learning and memory (Barco et al., 2002). The robust activation of CREB is achieved via activation of the Ras–ERK1/2 and the nuclear Ca²⁺–calmodulin (CaM) kinase pathways (Hardingham et al., 2001b). Because CaM kinase activity itself is Ca²⁺-dependent (Bading and Greenberg, 1991; Finkbeiner and Greenberg, 1996) this pathway mediates CREB phosphorylation within the first few seconds of Ca²⁺ influx. On the other hand, the more upstream position of the Ca²⁺ dependent step (Ras) in the ERK1/2 pathway results in a slower yet prolonged activation after cessation of the synaptic input. The resulting CREB phosphorylation on Ser¹³³ allows the recruitment of the transcriptional co-activator, CREB-binding protein (CBP) to the target promoter (Chrivia et al., 1993). The trans-activating potential of CBP is further positively regulated by phosphorylation of Ser³⁰¹ by either calmodulin (Deisseroth et al., 1998) or CaM kinase IV (Chrivia et al., 1993; Chawla et al., 1998; Hardingham et al., 1999) after elevation of nuclear Ca²⁺ (Impey et al., 2002). Growing evidence suggest that physiological levels of synaptic NMDA receptor activation may enhance resistance to trauma and promote survival of various neuronal types (Hardingham and Bading, 2010). Conversely, intense or chronic activation of extrasynaptic NMDA receptors has been implicated as the leading cause of neuronal death following acute traumas such as stroke, mechanical trauma, or seizure activity (Choi, 1990; Hardingham and Bading, 2010). Moreover, NMDA receptor activity is thought to contribute to the etiology of many chronic neurodegenerative disorders, such as Huntington’s disease, HIV-associated dementia, and AD (Lancelot and Beal, 1998; Chohan and Iqbal, 2006; Fan and Raymond, 2007).

AMPA receptors are heterotetrameric structures composed of subunits encoded by four genes GluR1–4, and primarily conduct Na⁺ and K⁺ currents. The permeability to Ca²⁺ is regulated by the presence of GluR2 subunits. Only the AMPA receptors lacking GluR2 are permeable to Ca²⁺, although with reduced affinity compared to NMDA receptors (Dingledine et al., 1999). The GluR2 lacking AMPA receptors are widely expressed in the CNS (including interneurons, stellate, and glial cells) where they contribute to synaptic transmission and changes in synaptic efficacy (Isaac et al., 2007), as well as induce multiple forms of synaptic plasticity, including LTP (Gu et al., 1996; Jia et al., 1996; Kullmann and Lamsa, 2007; Liu and Zukin, 2007). Similarly to the NMDA receptor, Ca²⁺-permeable AMPA receptors signal to the nucleus to activate CREB (Perkinton et al., 1999; Tian and Feig, 2006) and other transcriptions factors. In response to Ca²⁺ transients both CaMKII and PKA phosphorylate specific AMPA subunits modifying the synaptic transmission strength (Lee and Kirkwood, 2011). Direct phosphorylation regulates the opening probability of the channels, as well as the insertion of the receptors into the postsynaptic membrane. During LTP phosphorylation increases the opening probability and the concentration of AMPA receptors at the synapse, whereas phosphorylation and receptor density decrease during LTD (Malenka, 2003).

KARs respond to kainate, glutamate, and, with very low affinity, to AMPA (Hollmann and Heinemann, 1994). KARs are tetrameric combinations of five subunits: GluK1, GluK2, GluK3, GluK4, and GluK5 (Hollmann and Heinemann, 1994; Collingridge et al., 2009). The GluK1–3 subunits are highly homologous and can form functional receptors as homotetramers, while GluK4–5 share only 45% homology with GluK1–3 and must combine with any of them in order to generate functional KARs (Herb et al., 1992). The receptor structure consist of extracellular N-terminal and ligand binding domains implicated respectively in subunit and ligand recognition, a transmembrane region, and intracellular re-entrant loop (p loop), forming the Ca²⁺ pore and C-terminal regions (Hollmann et al., 1994; Wo and Oswald, 1994). Similarly to the GluR2 subunit of AMPA receptors GluK2 and GluK3 can undergo RNA editing resulting in changes from a conserved glutamine (Q) to an arginine (R) in the channel pore loop that completely abolish Ca²⁺ permeability (Egebjerg and Heinemann, 1993; Kohler et al., 1993). KARs functions are much less understood that those of AMPA and NMDA receptors. They are located both pre- and postsynaptically, and based on genetic and pharmacological studies they appear to act mostly as modulators rather than obligatory components of synaptic transmission and neuronal excitability (Chittajallu et al., 1999; Contractor et al., 2011). KARs can regulate both excitatory and inhibitory synaptic transmission (Rodríguez-Moreno et al., 1997; Contractor et al., 2000; Kamiya and Ozawa, 2000; Schmitz et al., 2000; Frerking et al., 2001; Jiang et al., 2001). In the hippocampus, at mossy fiber CA3 pyramidal neuron synapses, endogenous activation of presynaptic kainate receptors increases the probability of glutamate release (Schmitz et al., 2001; Huettner, 2003). A similar regulatory role on excitatory transmission has been observed for other types of synapses in the central and peripheral nervous system (Pinheiro and Mulle, 2006). In addition to facilitation, presynaptic activation of KARs has also been shown to lead to inhibition of glutamate release (Kamiya and Ozawa, 1998; Rozas et al., 2003; Lauri et al., 2005, 2006; Jin et al., 2006). KARs-mediated bidirectional modulation has also been observed for the inhibitory transmitter GABA. KARs stimulation enhances GABA release at synapses between interneurons (Mulle et al., 2000; Cossart et al., 2001), yet it inhibits GABA release at interneuron–pyramidal cell connections (Clarke et al., 1997; Rodríguez-Moreno et al., 1997; Rodríguez-Moreno and Lerma, 1998). These opposite effects seem to rely on the fact that KARs signaling can occur either via ion influx, or by coupling of the receptor to G-proteins and activation of PKC dependent mechanisms (Melyan et al., 2002; Rozas et al., 2003; Ruiz et al., 2005). Overall it is believed that ionotopic activity accounts for enhanced neurotransmitter release, while inhibition is triggered by the metabotropic signaling pathway. In addition to neurotransmitter release, KARs contribute to temporal summation of postsynaptic depolarization in response to bursts of action potentials (Castillo et al., 1997; Vignes and Collingridge, 1997; Frerking and Ohlinger-Frerking, 2002; Jin et al., 2006), modulation of short- and LTP, as well as LTD, thus to learning and memory (Bortolotto et al., 1999; Contractor et al., 2001; Lauri et al., 2003; Schmitz et al., 2003; Campbell et al., 2007), nociception (Ko et al., 2005), synaptogenesis, and neuronal development (Lerma, 2006; Lauri and Taira, 2011).

As per pathological scenarios, KARs are considered to contribute to the induction and propagation of seizures (Ben-Ari, 1985). Although unequivocal involvement of KARs in human epilepsies still has to be established, KAR mutant mice display altered susceptibility to kainic acid induced seizures (Mulle et al., 1998; Vissel et al., 2001), and selective agonism of GluK1 blocks pilocarpine-induced epileptiform activity in hippocampal slices and seizures in vivo (Smolders et al., 2002). Like other glutamate receptors, KARs have been implicated in a variety of neurodegenerative conditions where glutamate excitotoxicity is believed to contribute to neuronal cell death. For example, GluK1 antagonists afford good neuroprotection in focal and global ischemic models (Bullock et al., 1994; Gill and Lodge, 1994; O’Neill et al., 1998, 2000). GluK2 genotype variations, and possibly excitotoxic mechanisms, have been linked to variations in the age of onset of Huntington’s disease (Rubinsztein et al., 1997). Similar associations linking KARs with neuropsychiatric disorders have begun to emerge from various human genetic and post-mortem studies. KARs levels are reduced in bipolar and schizophrenic patients (Begni et al., 2002; Pickard et al., 2006; Wilson et al., 2006; Beneyto et al., 2007).

The human GluR6 KAR subunit, GRIK2, has been implicated in obsessive–compulsive disorder (Delorme et al., 2004), schizophrenia (Bah et al., 2004), and autism (Strutz-Seebohm et al., 2006). Human GluR7, GRIK3, and KA1 subunit, GRIK4, may be susceptible factors in major depressive disorders (Schiffer and Heinemann, 2007).

Nicotinic acetylcholine receptors

The nACh receptors are included in the cys-loop ligand-gated ion channel superfamily, and are activated by the endogenous neurotransmitter acetylcholine (ACh), as well as nicotine, hence the name. The receptors are pentameric homo or hetero combinations of different subunits types, α2–10, β2–4, with non-selective cation permeability and unique affinity to Ca²⁺ (Fucile, 2004). The nACh receptors are widely expressed in the brain at pre- and postsynaptic, as well as extrasynaptic loci (Dani and Bertrand, 2007). Presynaptic and pre-terminal nACh receptors enhance neurotransmitter release; postsynaptic nACh receptors contribute to fast excitatory transmission, while extrasynaptic nACh receptors influence neuronal excitability and/or intracellular processes (Dani and Bertrand, 2007). The nACh receptors play important modulatory roles in neuronal development and synaptic plasticity, participating in cognitive functions such as learning, memory, and attention (Levin and Simon, 1998; Mansvelder and McGehee, 2000; Sweatt, 2001). Inhibition of nACh receptors results in memory deficits, while the use of agonist has proved beneficial in different types of memory, such as short-term and working memory, both in animals and humans (Wallace and Porter, 2011). Furthermore, diminution, disruption, or alteration in the function of nACh receptors contributes to dysfunctions associated with various neurodegenerative pathologies, including epilepsy, Parkinson’s disease, Alzheimer’s disease, schizophrenia, autism, and addiction (Dani and Bertrand, 2007). The mechanisms underlying the effects of nACh receptors in synaptic plasticity are still poorly understood. In most cases nACh currents are thought to contribute to postsynaptic depolarization, enabling the activation of the VDCCs and subsequent Ca²⁺ influx, which augments the primary Ca²⁺ signals generated by the direct Ca²⁺ influx through nACh receptors (Dajas-Bailador et al., 2002). On the other hand, the predominant neuronal form, the homomeric α7 nACh receptor subtype has higher affinity for Ca²⁺ than Na⁺ (Shen and Yakel, 2009). Activation of α7 nACh receptors can thus generate Ca²⁺ transients via its channel pore sufficient to trigger Ca²⁺-induced Ca²⁺ release from ryanodine-dependent stores and downstream Ca²⁺-dependent intracellular processes independently of VDCCs activation (Sharma and Vijayaraghavan, 2001).

Serotonin receptor (5-HT₃)

5-HT₃ is the only serotonin receptor acting as a ligand-gated ion channels rather than a metabotropic receptor. Similarly to nACh it belongs to the family of cys-loop ligand-gated ion channel (Maricq et al., 1991). Within the CNS, 5-HT₃ receptors are highly expressed in brainstem areas, such as the area postrema and the nucleus of the solitary tract. Within the forebrain, 5-HT₃ receptors have been found in the entorhinal, frontal and cingulate cortices, hippocampus, and amygdala. Notably, despite clear pharmacological and physiological evidence that they affect dopamine neurotransmission, 5-HT₃ receptors are found at very low levels in areas such as striatum, substantia nigra, thalamus, or dorsal raphe nucleus (Kilpatrick et al., 1987; Tecott et al., 1993; Färber et al., 2004). 5-HT₃ receptor consist of five subunits with an extracellular N-terminal domain containing the ligand recognition site for serotonin, four transmembrane domains (T2–T4), and a short intracellular C-terminal (Maricq et al., 1991). T2 is responsible for the formation of the channel pore, which conducts primarily Na⁺- and K⁺- and Ca²⁺ (Yang, 1990; Hargreaves et al., 1994), while the large intracellular loop between T3 and T4 is the site of protein kinase phosphorylation and alternative splicing (Maricq et al., 1991; Davies et al., 1999; Dubin et al., 1999). Of the five different isoforms cloned to date, A–E, only 5-HT_3A can generate functional homomeric receptors, all others isoforms must heteropentamerize with 5-HT_3A (Maricq et al., 1991; Davies et al., 1999; Dubin et al., 1999). The best characterized receptors are 5-HT_3A and 5-HT_3AB, with the homomeric form displaying lower single channel conductance, higher Ca²⁺ permeability, and slower kinetics of activation, deactivation, and desensitization (Davies et al., 1999; Hayrapetyan et al., 2005). The 5-HT₃ receptors are expressed both pre- and postsynaptically. Presynaptic 5-HT₃ receptors display high permeability to Ca²⁺ (Nichols and Mollard, 1996; Nayak et al., 1999), whereas postsynaptic receptors display a lower permeability to Ca²⁺ compared to Na⁺ and K⁺ (Yakel et al., 1990; Yang, 1990). Presynaptic 5-HT₃ stimulation leads to opening of the ion channel, rapid membrane depolarization and release of various neurotransmitters, such as dopamine and GABA (Koyama et al., 2000; van Hooft and Vijverberg, 2000; Yakel, 2000). The activation of postsynaptic 5-HT₃ receptors contributes to fast excitatory synaptic transmission in various brain areas, such as the lateral amygdala (Sugita et al., 1992) and the visual cortex (Roerig et al., 1997). The best known physiological functions of 5-HT₃ receptors are the regulation of nausea and vomit (Fozard and Mobarok, 1978; Florczyk et al., 1982), and of mesocorticolimbic neurotransmission, which implicates these channels in the etiology of certain forms of drugs addiction (i.e., alcohol, cocaine; Engleman et al., 2008). In addition, modulation of 5-HT₃ receptors has also been suggested to have therapeutical relevance for schizophrenia, anxiety, cognition, and nociception (Thompson and Lummis, 2007).

ATP P2X receptors

The ionotropic P2X receptors are channels that respond to extracellular ATP to induce membrane depolarization and Ca²⁺ influx. P2X receptors are widely distributed pre- and postsynaptically on different cell types in the brain and spinal cord (Burnstock and Knight, 2004). Sensitivity to ATP, Ca²⁺ permeability, desensitization, recovery from desensitization kinetics, and other biophysical and pharmacological properties of the channels depend upon the subunit channel composition. Seven different subunits, P2x_1–7, can contribute to the generation of heterotrimers, or, with the exception of P2X6, homotrimers (Browne et al., 2010). P2X receptors display considerably high Ca²⁺ permeability, thus can mediate substantial Ca²⁺ influx despite the fact that the amplitude of P2X-mediated currents is quite modest when compared for example to glutamate-evoked responses (Pankratov et al., 2008). Because of this, P2X are the main postsynaptic Ca²⁺ entry channels at resting potential when NMDA receptors are blocked by Mg²⁺ (Pankratov et al., 2003). P2X receptors can dynamically interact with other neurotransmitter receptors, including NMDA receptors, GABA_A receptors, and nACh receptors (Surprenant and North, 2009). Activation of P2X receptors has multiple modulatory effects on synaptic plasticity, either inhibiting or facilitating the long-term changes of synaptic strength depending on the physiological context; however the precise mechanisms of P2X-dependent regulation of synaptic plasticity remain elusive (Pankratov et al., 2008).

Intracellular Calcium Stores

It is well established that in addition to intracellular influx, global Ca²⁺ signals comprise release from intracellular stores either as a result of Gq-coupled receptor activation, or secondary to the [Ca²⁺]_i rise itself. The largest intracellular store able to accumulate Ca²⁺ to concentrations of 10–100 mM is the endoplasmic reticulum (ER). ER Ca²⁺ release is mediated by RyRs, and by IP₃Rs. The function of the ER Ca²⁺ channels is to amplify or trigger Ca²⁺ rises initiated by the plasmalemma Ca²⁺ influx. In addition to the ER, mitochondria have also been shown to act as intracellular Ca²⁺ stores and play a prominent role in determining the shape, amplitude, and duration of the Ca²⁺ transients.

Endoplasmic reticulum calcium regulation

In neurons the ER is represented by a complex system of folded membranes extending from the nuclear envelope throughout perikaryon, axon (Henkart et al., 1978), presynaptic terminals (McGraw et al., 1980), dendrites, and dendritic spines (Satoh et al., 1990). Since the first demonstrations of its Ca²⁺ sequestering ability in squid giant axon (Henkart et al., 1978) and rat synaptosomes (McGraw et al., 1980), the ER has been the object of intensive studies in order to determine its role in neuronal Ca²⁺ homeostasis (Andrews et al., 1988; Markram et al., 1995). It is now evident that ER Ca²⁺ regulation plays important functions in neuronal physiology and plasticity regulating synaptic transmission both at the presynaptic and postsynaptic terminals (Berridge, 1998; Park et al., 2008). Notably, at the level of dendritic spines, the ER comes in contact with the postsynaptic density thanks to the protein Homer simultaneously binding IP₃Rs on the ER, and group 1 metabotropic receptors at the plasma membrane (Tu et al., 1998). Similar juxtaposition are also observed in other areas of the neuron and are believed to regulate extracellular Ca²⁺ entry through store operated channels or membrane excitability via activation of K⁺ channels (Benedeczky et al., 1994; Berridge, 2002). ER Ca²⁺ homeostasis relies upon Sarco/endoplasmic-reticulum Ca²⁺ ATPase (SERCA) pumps and binding proteins regulating respectively Ca²⁺ uptake from the cytosol and Ca²⁺ sequestration (Prins and Michalak, 2011; Vandecaetsbeek et al., 2011). Conversely, Ca²⁺ efflux from the ER is achieved though input activated IP₃Rs and RyRs release or Ca²⁺ leak. The importance of the ER Ca²⁺ homeostasis in brain physiology can be easily appreciated considering the number of neurodegenerative disorders have been linked to alterations with its various components (Mattson, 2007).

Ryanodine receptors

To date, three mammalian isoforms of RyRs have been isolated displaying unique differential expression in the CNS. RyR1 is found exclusively in Purkinje cells in the cerebellum (Furuichi et al., 1994), RyR2 is highly expressed in Purkinje cells and the cerebral cortex (Sharp et al., 1993; Furuichi et al., 1994), while RyR3 has a wider distribution and is also present in hippocampus, thalamus, and striatum (Hakamata et al., 1992). RyRs are homotetramers and the largest known ion channels with a mass greater than 2 MDa, a fact that have challenged the analysis and molecular understanding of their function. Nevertheless, in recent years it has been shown that due to their close physical interaction with Ca_v1.1–1.2 L-type Ca²⁺ channels, RyRs directly support Ca²⁺-induced Ca²⁺ release (CIRC) as consequence of Ca²⁺ influx through VDCCs or ROCs (Chavis et al., 1996). Cytosolic and luminal Ca²⁺ levels are the principal direct and indirect regulators of RyRs functions (Meissner, 2002). RyRs have multiple cytosolic allosteric Ca²⁺ binding sites, and in absence of other effectors are maximally activated at cytosolic Ca²⁺ concentrations of 1–10 μM, while they are inhibited in the low millimolar range (Bezprozvanny et al., 1991). Luminal Ca²⁺ content has also been shown to modify the availability and open probability of RyRs (Shmigol et al., 1996), possibly by both direct interaction with luminal Ca²⁺ regulatory sites (Sitsapesan and Williams, 1995), and by binding to the cytoplasmic sites after permeating through the pore (Tripathy and Meissner, 1996; Laver, 2007). RyRs-dependent Ca²⁺ release is also potentiated by ATP and other adenine nucleotides (Meissner, 1984). Furthermore, as components of a regulatory macromolecular complex with PKA, FKBP12, FKBP12.6, calsequestrin, triadin, and junction, calmodulin and CaMKII are responsible for most of the indirect effects of Ca²⁺ on RyRs. Calmodulin binds all three forms of RyRs, both in its Ca²⁺-free (ApoCaM) and Ca²⁺-bound forms (CaCaM; Tripathy et al., 1995; Yamaguchi et al., 2005). ApoCaM is a partial agonist whereas CaCaM is an inhibitor (Rodney et al., 2000). In Drosophila melanogaster, CaMKII activation and RyRs-dependent Ca²⁺ release are fundamental for post-tetanic potentiation of neurotransmitter secretion (Shakiryanova et al., 2007). In addition to neurotransmitter release (Mothet et al., 1998), RyRs also regulate action potential hyperpolarization (Kawai and Watanabe, 1989) and axonal retrograde transport (Breuer et al., 1992). In hippocampus (Reyes and Stanton, 1996) and in Purkinje cells (Kohda et al., 1995), LTD has been shown to be RyRs-sensitive Ca²⁺ stores dependent. Despite the evidence suggesting a potential role in synaptic plasticity behavioral testing in RyR knock out mice have given discrepant results. Balschun et al. (1999) observed equivalent learning ability in RyRs knock out and wild-type animals. However, in the same knock out model others have reported facilitated CA1 LTP induction after short tetanic stimulation, absence of LTD and improved spatial ability (Futatsugi et al., 1999), as well as impairments of performance in the contextual fear conditioning test, passive avoidance test, and Y-maze (Kouzu et al., 2000).

Inositol-1,4,5-triphosphate receptors

Following activation of Gq-coupled or tyrosine kinase receptors by their ligands (i.e., hormones, growth factors, neurotransmitters) the activation of phospholipases leads to the cleavage of phosphatidylinositol 4,5-biphosphate and generation of second messengers diacylglycerol and inositol-1,4,5-triphosphate (IP₃) that can easily diffuse from the plasma membrane through the cell. Unique amongst the various cellular second messengers IP₃ binds and activates an intracellular ligand-gated Ca²⁺ channel, IP₃R. In mammals the receptor is an homo- or hetero-tetramer formed by the product of three IP₃R genes (IP₃R1–3). Each gene is translated as several splicing variants, and its expression can be modulated by various stimuli providing an impressive level of channel diversity (Foskett et al., 2007). Like RyRs the cytosolic portion of the IP₃Rs provides scaffold to a host of regulatory proteins creating macromolecular complexes able to receive inputs from the majority of the signaling pathways, as well as to sense metabolic changes within the cell. The various receptor subtypes display different binding affinities for IP₃ with IP₃R2 being more sensitive than IP₃R1, and both considerably more sensitive than IP₃R3 (Iwai et al., 2005). The binding of IP₃ not only controls channel opening, but it is also necessary for its time dependent inactivation (Mikoshiba, 2007), and clustering of the receptors (Taylor et al., 2009). IP₃R-evoked Ca²⁺ signals can operate locally or globally thanks to the hierarchical recruitment of elementary Ca²⁺ release events, and ability to form clusters. The smallest event known as “Ca²⁺ blip” is likely the result of the random opening of single IP₃R, last about 130 ms and causes very small increases of cytosolic Ca²⁺ (<50 nM). Ca²⁺ puffs are larger (50–600 nM) and longer lasting events spreading few micrometers, and are caused by the near-simultaneous opening of several clustered IP₃Rs (Bootman and Berridge, 1996). If the inducing stimulus persist the frequency of Ca²⁺ puffs increases, and can originate regenerative Ca²⁺ waves as the Ca²⁺ diffusing from one site ignites the activity of another receptor (Berridge, 1997). In addition to IP₃ the activation of the receptor requires Ca²⁺ as co-agonist (Finch et al., 1991). It is well established that cytosolic Ca²⁺ regulates IP₃Rs activity in a biphasic fashion, with isotype differences in the stimulatory and inhibitory ranges (Foskett et al., 2007; Mikoshiba, 2007). While it is overall agreed that modest increases in cytosolic Ca²⁺ enhance responses to IP₃ and higher concentrations inhibit it, the modalities of Ca²⁺ regulation are still unclear. Both direct regulation via binding to stimulatory or inhibitory sites on the channel structure, and indirect regulation through one of the accessory regulatory proteins have been proposed (Taylor et al., 2004). Luminal levels of Ca²⁺ are also known to impact gating of IP₃Rs via mechanisms involving Ca²⁺ binding and interaction with chaperones such as calnexin and HRp44 (Higo et al., 2005). As seen with other Ca²⁺ channels, IP₃R activity is as well modified by phosphorylation (Foskett et al., 2007), ATP levels (Mak et al., 1999, 2001), redox status (Higo et al., 2005), and interaction with other proteins (Patterson et al., 2004; Foskett et al., 2007). One significant example is the interaction between IP₃R1 and the voltage-dependent anion channels (VDACs) located in the mitochondria outer membrane (Szabadkai et al., 2006), which facilitates mitochondrial Ca²⁺ uptake. Over the past decade, thanks to the generation of IP₃R knock out mice models it has been established that IP₃Rs play a crucial role in brain functions, enhancing neurotransmitter release following repetitive stimulation (Emptage et al., 2001), neurite formation and extension during development (Takei et al., 1998), hippocampal LTP (Fujii et al., 2000), cerebellar LTD (Inoue et al., 1998), learning, memory, and behavior (Matsumoto et al., 1996; Fujii et al., 2000; Nishiyama et al., 2000).

Presenilins as endoplasmic reticulum Ca²⁺ leak channels

The steady state Ca²⁺ within the lumen of the ER is maintained by balancing the influx created by SERCA pumps with “leak” mechanisms. In neurons, the addition of SERCA pumps inhibitors results in an immediate drop of the luminal Ca²⁺, implying the existence of leak channels (Solovyova et al., 2002; Verkhratsky, 2005). In muscle cells it has been shown that RyRs under conditions causing their uncoupling from the luminal Ca²⁺ binding proteins can function as leak channels (Mark et al., 1998, 2000). The molecular identity of the ER leak channels under physiological conditions is still unresolved. However, recent evidence points to presenilins (PSs) as potential physiological ER Ca²⁺ leak channels (Tu et al., 2006; Nelson et al., 2007). Presenilin 1 and 2 are integral ER transmembrane proteins (Annaert et al., 1999) undergoing endoproteolytic cleavage that generates N-terminal and C-terminal fragments. The cleaved PSs fragments assemble in a complex with nicastrin, anterior pharynx defective 1, and presenilin enhancer 2, translocate to the plasma membrane where they function as the γ-secretase enzyme responsible of the amyloid precursor protein (APP) cleavage generating amyloid β, the primary component of amyloid plaques in AD (De Strooper et al., 1998; De Strooper and Annaert, 2010). Since the discovery that PSs mutations account for the majority of familiar forms of AD, PSs have been intensively studied (Levy-Lahad et al., 1995; Rogaev et al., 1995; Sherrington et al., 1995). Together with the understanding of their role in the APP processing come the evidence that they participate to ER Ca²⁺ homeostasis. Evidence of deregulated ER Ca²⁺ signaling has been gathered from various experimental systems including human fibroblasts from AD patients and transgenic mouse models carrying PSs mutations (Ito et al., 1994; Guo et al., 1996, 1999; Leissring et al., 1999a,b; Stutzmann et al., 2004, 2006). The mechanisms proposed to explain PSs functions in Ca²⁺ homeostasis are based on physical interactions and indirect influence on others ER Ca²⁺ players, such as RyRs (Chan et al., 2000; Stutzmann et al., 2006), IP₃Rs (Cai et al., 2006; Cheung et al., 2008), SERCA pumps (Green et al., 2008), as well as on the influence on store operated Ca²⁺ influx (Leissring et al., 2000). However, recent experiments using planar lipid bilayers and neurons from double knock out PS1/PS2 and triple transgenic AD mice, showed that PSs themselves form low conductance Ca²⁺ channels responsible for about 80% of the passive Ca²⁺ leak from the ER (Tu et al., 2006; Nelson et al., 2007). This ER Ca²⁺ leak function appears to be independent from the γ-secretase activity, and is lost in most of the familiar AD PSs mutants studied to date resulting in ER Ca²⁺ overload (Tu et al., 2006; Nelson et al., 2007; Bezprozvanny and Mattson, 2008). Regardless of the type of mechanisms, a role for PSs in synaptic plasticity, learning, and memory via regulation of Ca²⁺ signaling has been well established. PSs regulates homeostatic synaptic scaling (Pratt et al., 2011), neurotransmitter release from the presynaptic terminal (Zhang et al., 2009), LTP (Zhang et al., 2010), and the establishment of hippocampal memory (Saura et al., 2004).

Mitochondria and calcium

Thanks to their ability to rapidly uptake large quantities of Ca²⁺, mitochondria contribute to shaping the amplitude and duration of cytosolic Ca²⁺ signaling (Deluca and Engstrom, 1961; Vasington and Murphy, 1962). Structural dynamic junctions link the mitochondria to the ER, facilitating the signal transduction, and biosynthetic interplay existing between the two organelles (Csordás et al., 2006; Hayashi et al., 2009; Rizzuto et al., 2009; de Brito and Scorrano, 2010). Amongst the proteins identified as possible components of the ER–mitochondria coupling complexes are glucose-related protein 75 (Szabadkai et al., 2006), sigma-1 receptor (Hayashi and Su, 2007), mitofusin-2 (de Brito and Scorrano, 2009), phosphofurin acidic cluster sorting protein 2 (Simmen et al., 2005), BAP-31 (Wang et al., 2000), and dynamin like protein 1 (Pitts et al., 1999). These interactions regulate the trafficking of phospholipids, as well as the Ca²⁺ interchanges between the ER and the mitochondria. Ca²⁺ can cross both the outer and inner mitochondrial membranes to enter the matrix, using different pathways. The mitochondrial Ca²⁺ influx is a tightly controlled process, because under physiological concentrations Ca²⁺ modulates pyruvate, isocitrate, and α-ketoglutarate dehydrogenases, thus the Krebs cycle (Denton et al., 1980; Hansford, 1980) and the production of ATP (Jouaville et al., 1999). The predominant mechanism of intake through the outer membrane is represented by high conductance and high-density VDACs channels (McEnery et al., 1993; Lee et al., 1998). The inner membrane Ca²⁺ import is primarily regulated through the mitochondrial Ca²⁺ uniporter (MCU) composed by a protein forming the channel itself (Baughman et al., 2011; De Stefani et al., 2011), and the accessory protein MICU1 regulating its function (Perocchi et al., 2010). The MCU allows the ion transport down its electrochemical gradient and it is functional only at high extra mitochondrial Ca²⁺ concentrations. Rapid Ca²⁺ intake independent from the electrochemical gradient is achieved through mitochondrial RyR (Beutner et al., 2001), and the rapid mode mitochondrial Ca²⁺ transport (Sparagna et al., 1995). Both systems operate at low physiological Ca²⁺ levels while they are inactivated at concentrations causing the activation of MCU. Ca²⁺ efflux is mostly dependent on the mitochondrial Na⁺/Ca²⁺ exchanger that couples Ca²⁺ extrusion with the inward directed Na⁺ electrochemical gradient (Carafoli et al., 1974; Palty et al., 2010), and possibly by the transient openings of the permeability transition pore (PTP). The high conductance channel is a complex of yet non-identified molecular components, that once opened allows indiscriminate exchange of ions and solutes between the cytosol and the mitochondrial matrix. VDACs on the outer membrane, adenine nucleotide transferase on the inner one, and cyclophilin D in the matrix, have been proposed as possible channel components, but studies from knock out mice have raised doubts on their role as obligatory constituents of the PTP (Rizzuto et al., 2009). The opening of the PTP channel is induced by high matrix Ca²⁺, free radicals, adenine nucleotide depletion, pH, and decreased mitochondrial membrane potential (Halestrap, 2009). The process is reversible, yet under non-physiological conditions the PTP can become fixed in the open conformation leading to apoptotic cell death via release of apoptotic factors such as cytochrome c, apoptosis-inducing factor and Smac/Diablo, or necrotic cell death via massive ATP depletion (Abou-Sleiman et al., 2006; Leung and Halestrap, 2008). Finally the Ca²⁺/H⁺ uniporter Letm1 can move Ca²⁺ in or out of the mitochondria in a Ca²⁺ and pH-gradient dependent manner (Jiang et al., 2009).

Neuronal synapse are highly enriched with mitochondria in order to fulfill the high energy requirements needed to fuel the active processes required for synaptic transmission, and to serve as Ca²⁺ sinks and modulators following Ca²⁺ entry (David and Barrett, 2000; Tsang et al., 2000). The local ATP production by mitochondria alone is important to maintain synaptic transmission. Impairment of mitochondria transport into the presynaptic terminal have been shown to cause abnormal neurotransmission during intense stimulation (Verstreken et al., 2005), impaired presynaptic short-term plasticity, and accelerated synaptic depression under high-frequency firing (Ma et al., 2009). Furthermore, genetically engineered mice with decrease mitochondrial respiratory function displayed altered retention and consolidation of spatial memory (Tanaka et al., 2008). As per the Ca²⁺ regulatory function, mice lacking mitochondrial VDACs show impaired fear conditioning and spatial learning, and deficits in long and short-term hippocampal synaptic plasticity (Weeber et al., 2002). Tetanus-induced synaptic potentiation, contrary to other forms of synaptic plasticity, does not require Ca²⁺ influx into the cell, but rather depend on Na⁺ influx into the nerve terminals to promote Ca²⁺ efflux from mitochondria through the Na⁺/Ca²⁺ exchanger (Tang and Zucker, 1997; Yang et al., 2003). Additionally, while the majority of synapses requires the cooperation between ER and mitochondria to regulate Ca²⁺ signaling (Rizzuto et al., 2004), mitochondria appear to be the main Ca²⁺ regulating system, with minimal uptake into the ER during substained prolonged stimulation in specialized structures, like the neuromuscular junction (David and Barrett, 2003) and the calyx of Held (Billups and Forsythe, 2002).

Calcium Aging and Neurodegeneration

The plasticity of the nervous system depends at any time point on the balance between degenerative and regenerative processes. Because Ca²⁺ is a fundamental signaling mechanism involved in almost all cellular physiological functions, subtle alterations of its homeostasis lead to profound functional changes. Several lines of evidence support the notion that Ca²⁺ dyshomeostasis is implicated in normal brain aging (Gibson and Peterson, 1987; Disterhoft et al., 1994; Khachaturian, 1994). The cognitive decline occurring with normal aging is not associated with significant neuronal loss (Gallagher et al., 1996), but is rather the result of changes in synaptic connectivity. Age-dependent alterations have been observed for multiple components the Ca²⁺ cellular machinery, and appear to strongly correlate with cognitive deficits (Power et al., 2002; Rosenzweig and Barnes, 2003). Aged hippocampal neurons display decreased synaptic plasticity with increased Ca²⁺ influx and density of L-type VDCCs (Campbell et al., 1996; Thibault and Landfield, 1996; Thibault et al., 2001). The increase could arise from altered gene or protein expression (Herman et al., 1998), or phosphorylation changes of the L-type Ca²⁺ channels (Norris et al., 2002; Davare and Hell, 2003). GluR2 and NR1 protein expression is reduced in hippocampus (Liu et al., 2008; Yu et al., 2011) and postrhinal and entorhinal cortex of aged rats (Liu et al., 2008). Additionally, aged neurons show enhanced CIRC due to the increased ER Ca²⁺ release (Kumar and Foster, 2004; Gant et al., 2006), diminished Ca²⁺ extrusion through the plasma membrane ATPase (Michaelis et al., 1996; Gao et al., 1998), reduced cellular Ca²⁺ buffering capacity due to impairment of the SERCA pumps (Murchison and Griffith, 1999), and diminished mitochondrial Ca²⁺ sink capability (Xiong et al., 2002; Murchison et al., 2004), activation of calcineurin (Foster et al., 2001), and calpains (Nixon et al., 1994). The overall result is an increase of Ca²⁺ loads which negatively impact neuronal Ca²⁺-dependent K⁺ channel slow after-hyperpolarization and excitability (Landfield and Pitler, 1984; Khachaturian, 1989; Matthews et al., 2009), increases the threshold frequency for induction of LTP (Shankar et al., 1998; Ris and Godaux, 2007), enhances the susceptibility to induction of LTD (Norris et al., 1996; Kumar and Foster, 2005; Lee et al., 2005), and ultimately learning and memory. The idea that Ca²⁺ dyshomeostasis is a key factor in determining brain aging is substantiated by various studies using pharmacological interventions aimed at counteracting the age-related Ca²⁺ signaling increase (Foster, 2006). Administration of BAPTA-AM, a membrane-permeant Ca²⁺ chelator, ameliorates impaired presynaptic cytosolic, and mitochondrial Ca²⁺ dynamics in hippocampal CA1 synapses of old rats (Tonkikh and Carlen, 2009), and enhances spatial learning (Tonkikh et al., 2006). Similarly, the L-type Ca²⁺ channel blocker nimodipine counteracts age-related learning impairments in rabbits (Deyo et al., 1989; Kowalska and Disterhoft, 1994), rodents (Levere and Walker, 1992; Ingram et al., 1994), non-human primates (Sandin et al., 1990), and elderly patients with dementia (Ban et al., 1990; Tollefson, 1990). Aging is the greatest risk factor for neurodegenerative disorders, a heterogenous group of pathologies characterized by the gradual neuronal loss in motor, sensory, or cognitive systems. While per se not cause of neuronal loss, the age-dependent alterations of Ca²⁺ signaling can possibly enhance neuronal vulnerability to metabolic and functional stressors thus contribute to the initiation or progression of the neurodegenerative process (Toescu et al., 2004; Toescu and Vreugdenhil, 2010). Despite intrinsic different etiologies, dysregulated Ca²⁺, and mitochondrial homeostasis have emerged as common underlying molecular mechanisms of neuronal loss in Alzheimer’s, Parkinson’s, Huntington’s diseases, amyotrophic lateral sclerosis and other neurodegenerative disorders (Mattson, 2004, 2007; Bezprozvanny, 2009). The specificity of Ca²⁺ homeostatic and signaling machineries requirements that underlie the unique responses to the same stimuli of different neurons, accounts, at least partially, for the selective impairment of neuronal subtypes and brain areas observed during aging and neurodegeneration. While the sequence of pathological events and the kinetics of degeneration are still not completely understood, and likely differ amongst the various disorders, decreased mitochondrial functional capacity with diminished ATP production (Navarro, 2004) and increased reactive oxygen species generation (Floyd and Hensley, 2002), together with mitochondrial reduced Ca²⁺ buffering ability (Xiong et al., 2002), and enhanced Ca²⁺ responses are common key elements to ignite the necrotic or apoptotic cell death distinctive of most neurodegenerative disorders (Green and Kroemer, 2004; Lin and Beal, 2006).

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.